invasimapr 0.2.0

Release notes for invasimapr v0.2.0

🌿 A maturity & standards release for invasimapr

This release brings invasimapr in line with the B-Cubed software development guide, hardens the codebase, and overhauls the documentation — making the full traits → competition → invasion-fitness workflow easier to install, run, and cite. R CMD check now passes cleanly.

✨ Highlights

• Runnable quick-start — the README now ships a self-contained, copy-paste example that runs the pipeline end-to-end on the bundled demo data, from prepared inputs to species invasiveness and site invasibility.
• Standards-aligned — restructured to follow the B-Cubed guide, with a new Get started vignette and clean b3doc / pkgdown documentation builds.
• Easier to cite — archived on Zenodo with a citable concept DOI (10.5281/zenodo.20842472) and machine-readable CITATION.cff, codemeta.json and .zenodo.json metadata.

📚 Documentation

• Rewrote the README around a minimal, reproducible workflow, with manuscript context and key figures.
• Moved the step-by-step tutorials into pkgdown Articles (introduction, step-by-step workflow, clustering & risk scenarios, computing invasion fitness, and invasion-fitness synthesis) and added a Get started vignette.
• Renamed the bundled inputs_vignettes.rds to invasimapr_vignettes.rds.
• Added a Darwin Core-aligned data dictionary at inst/extdata/data_dictionary.csv.

🧹 Code quality

• Resolved the R CMD check WARNING and NOTEs (moved sp to Suggests, qualified glmmTMB::ranef, fixed switch(EXPR=), Rd line widths, and the CODE_OF_CONDUCT build ignore).
• Slimmed the install from ~100 MB to ~56 MB — gzipped the demo CSV, dropped six unused extdata files, and moved the vignette cache out of the build.
• Removed dead code (sigma_mat_from_vcov) and a redundant magrittr dependency.
• Declared scales and stringr as explicit imports; replaced the deprecated citEntry() in inst/CITATION with bibentry().
• DESCRIPTION now uses Title Case, with the maintainer ORCID and a copyright holder (Stellenbosch University).

⚙️ Compatibility

• Now requires R (≥ 4.1.0) — the package uses the native pipe |>.

🧰 The workflow at a glance

Seven high-level wrappers span the pipeline — prepare_inputs(), prepare_trait_space(), model_residents(), learn_sensitivities(), predict_invaders(), predict_establishment() and summarise_results() — built on core functions including compute_invasion_fitness(), compute_establishment_probability() and summarise_invasiveness_invasibility(). compute_invasion_fitness() gains an opt-in standardise_inputs argument to re-standardise raw predictors onto a common z-scale before computing fitness (off by default).

🔜 Roadmap

• Additional metrics for functional trait dispersion and environmental filtering.
• Interactive visualisation tools and scenario exploration modules.
• Expanded vignettes and tutorials.

---

💬 Questions or bugs? Please use the GitHub issue tracker.

invasimapr is developed within the B-Cubed project (Biodiversity Building Blocks for policy), funded by the European Union's Horizon Europe programme (grant 101059592), and is designed to pair with dissmapr for biodiversity data preparation.